Nemotron 3.5 ASR Streaming 0.6b (Concurrency)

Scoring

Primary metric: real_time_factor · Tie-breakers: throughput, gpu_pct

Metrics: real_time_factor, throughput, cpu_pct, ram_mb, gpu_pct, energy_wh, temperature_c

Changelog

• v1.0.0 — Initial profile (ADR-0012 shape, ADR-0013 engine). concurrency ceiling 8 set from measured per-instance VRAM, not copied from another engine.

Full definition

{
  "id": "nemotron-3-5-concurrency",
  "version": "1.0.0",
  "title": "Nemotron 3.5 ASR Streaming 0.6b (Concurrency)",
  "benchmark_type": "concurrency",
  "runtime": {
    "name": "nemotron",
    "min_version": "5.13.0"
  },
  "model": {
    "name": "nemotron-3.5-asr-streaming-0.6b"
  },
  "configuration": {
    "threads": 4,
    "concurrency": 1,
    "duration_s": 30
  },
  "scoring": {
    "primary_metric": "real_time_factor",
    "tie_breakers": [
      "throughput",
      "gpu_pct"
    ]
  },
  "metrics": [
    "real_time_factor",
    "throughput",
    "cpu_pct",
    "ram_mb",
    "gpu_pct",
    "energy_wh",
    "temperature_c"
  ],
  "overridable": {
    "concurrency": {
      "range": {
        "min": 1,
        "max": 8
      }
    },
    "threads": {
      "range": {
        "min": 1,
        "max": 16
      }
    }
  },
  "changelog": [
    {
      "version": "1.0.0",
      "notes": "Initial profile (ADR-0012 shape, ADR-0013 engine). concurrency ceiling 8 set from measured per-instance VRAM, not copied from another engine."
    }
  ]
}